Rains 2014; Muddiman and Stroud 2017; Rowe 2015; Sobieraj and Berry 2011). Another ... and Street 2007; Stroud et al. ...... Schwaber, Ken, and Mike Beedle.
Constructive comments? Designing an online debate system for the Danish Broadcasting Corporation This article brings together two strands of research that have potential to inform the development of constructive forms of journalism: online comments and media design. Through a three-year long case study of the development of new formats for online comments on the website of the Danish Broadcasting Corporation I explore the challenges encountered from two perspectives: the design features of the commenting system, and the design process. While the broadcaster has emphasized developing features for strengthening editorial control, user engagement has faltered. A lack of attention to users in the design process seems to have contributed to the problems. These findings have implications for constructive journalism's ambitions to facilitate audience engagement, in particular when tied to online platforms. Keywords: constructive journalism; design; human-computer interaction; interaction design; media design; online comments; participatory journalism; user experience;
Introduction Constructive journalism shares a number of characteristics with a seemingly quite different field, design: Both fields are oriented towards practice, towards solutions and towards the future. This article explores the interplay between design and journalism through a case study of a project at the Danish Broadcasting Corporation (DR), aiming to improve the systems for reader comments on the broadcaster’s website. In recent years, audience involvement through online comments has become a nearly ubiquitous, but also a highly contentious and often disparaged form of participatory journalism, as documented by Singer et al. (2011). However, the relation between constructive journalism, on the one hand, and online comments and other forms of participatory journalism on the other, is underexplored. Haagerup’s influential book on Constructive News is permeated by references to public debate, and presents itself as "a handbook of inspiration on how we can do better in the newsrooms, in the public debate and in our democracies" (2014, 16). However, except for a passing reference to a televised debate in which viewers were "adding their own contributions to the debate on the Internet" (Haagerup 2014, 132), he does not spend much attention on audience involvement in debate. McIntyre explicitly excludes participatory journalism from her definition, stating that "constructive journalism is different in that it does not have (...) a specific desire to encourage regular citizens to contribute content" (McIntyre 2015, 12). However, this seems somewhat at odds with her strong interest in effecting behavioral change in the audience: "A primary goal of constructive journalism is to engage news audiences. (...) This engagement might take place in the form of seeking more information, sharing stories on social media, signing a petition or donating time or money to a cause" (McIntyre 2015, 36). It seems unfortunate to a priori rule out that constructive journalism may also play a role in the facilitation of audience engagement online. In particular, given the current controversies surrounding the role of Facebook and other social media platforms in democratic debate, constructive journalism approaches might do well to include an ambition to not just inspire audience engagement, but also to facilitate it through platforms that are designed and maintained with attention to journalistic ideals and editorial responsibility. As Reich points out, online comments are both the most common and one of the most popular ways to facilitate audience participation in news (2011, 96–98). In a recent example, the influential newspaper the Guardian (31 January 2016) announced changes in their online commenting policies, stating that their goal was to host "constructive debate" and tying this ambition 1
to the ideals of public journalism. However, while a large body of journalism research has been devoted to studying the problems associated with online comments, it is striking to note that a design perspective on these challenges remains conspicuously absent, for example from Reich’s overview of research on online comments (2011, 100). In recent years, some media scholars have argued for approaches to media research that explore how academic theory can be made constructive, in the sense that it is applied to the design of new media forms. Such approaches have been presented using cross-disciplinary compound phrases such as “media design” (Koskinen 2006; Lunenfeld 2004; Løvlie 2011b), “communication design” (Morrison 2011; Skjulstad 2008), “aesthetic design” (Bolter, Engberg, and MacIntyre 2013) or simply "design interventions" (Löwgren and Reimer 2013a). Nyre et al. (2012) have championed media design as an approach to journalism research. Bolter has described this turn towards design as "a fusion of the critical stance of cultural theory with the constructive attitude of the visual designer" (2003, 30). The "constructive attitude" that is aimed at in these design approaches differs from the field of constructive journalism in some regards: while constructive journalism is rooted in journalism practice and mostly engages with the journalistic discourse, these design approaches tend to be positioned as academic research, and aim to actually construct solutions. However, the parallels between these two fields are stronger than they may seem. First of all, in contrast to other design approaches media design engages not just with the creation of systems and interfaces, but also with the production of texts (Fagerjord 2015; Løvlie 2011a) including news journalism (Øie 2015; Nyre et al. 2012). For both fields, the turn towards constructiveness implies a departure from ideals of "impartiality" or "objectivity", towards an explicitly acknowledged normative stance (see for instance Bardzell and Bardzell 2015; McIntyre 2015). There is also a profound reorientation from focusing on the past, towards seeking solutions for the future: To design, according to one seminal definition, is to "devise courses of action aimed at changing existing situations into preferred ones" (Simon 1969, 111); Krippendorff calls this "constructive interventions" (2006, 30). Given that the development of new forms of journalism is ever more closely linked to technological development and innovation, it is pertinent to explore the ways in which media design and constructive journalism may inform each other. R. Coleman has discussed the role of visual design in civic journalism (2007). However, the media design approaches cited above have less to do with visual design and more to do with approaches founded in information technology such as interaction design and human-computer interaction (HCI). The relation to these fields has so far gone largely unexplored in constructive journalism research. In what follows I will use interviews and observations of the DR project to discuss the following research question: RQ: How can insights from the design disciplines inform the development of constructive forms of journalism, especially with regards to online comments?
When questions of design are brought into discussions in other academic fields, there is a potential for confusion due to the double meaning of the word design: One, a noun describing the features of a system; and two, a verb referring to an activity and a process – designing (cf. Krippendorff 2016). The former meaning is probably the one that most non-designers tend to be interested in when speaking about design, and indeed when design is mentioned in media and journalism research about online comments, it usually refers to the features of the commenting system (see for instance Birchall and Coleman 2015; Singer et al. 2011). However, design researchers in interaction design and HCI tend to spend much attention on the second meaning, design as a process. In the following, I will attend to both these meanings of design, with an interest in seeing whether the design process may shed any light on the features of the designed system – and thereby point to implications for future endeavors in constructing systems for online comments.
2
Online comments Reader comments on mass media websites have long been a topic of contention both among media professionals, scholars and the public at large. While some scholars have been optimistic about the democratic potential of online comments, others have raised concerns about their deliberative value (Dahlgren 2005; Kies 2010; Sunstein 2017). Controversies regarding online comments have been heightened in recent years. Recently WIRED summarized a trend towards removing online comments from a number of popular websites, declaring it "the End of the Comments" (8 Oct 2015). Recent political events such as the “Brexit” referendum in the UK and the 2016 US presidential elections have raised fears about echo chambers and the reliability of online information ("fake news") high on the international public's agenda. In the Nordic countries, events such as the 2011 terror attack in Oslo and the 2015 refugee crisis have prompted public backlashes against online comments in Norway and Finland (Ahva and Hautakangas 2017; Ihlebæk, Løvlie, and Mainsah 2013; Løvlie, Ihlebæk, and Larsson 2017). Research on online comments in the fields of journalism, media and communication has often explored the quality of debate, as well as the challenge of incivility and harassment (Coe, Kenski, and Rains 2014; Muddiman and Stroud 2017; Rowe 2015; Sobieraj and Berry 2011). Another strand of research has explored what motivates and triggers participation (Chung 2008; Larsson 2011; Lee and Tandoc 2017; Ziegele et al. 2017). A recent study indicates that commenters are motivated by “the desire to interact” with journalists and to discuss with other users; however, the commenters "do not obtain cognitive gratifications to the desired extent" (Springer, Engelmann, and Pfaffinger 2015). Some research has also focused on the influence of system features and editorial policies on the deliberative quality of comments (Canter 2013; Ruiz et al. 2011; Toepfl and Litvinenko 2017; Wright and Street 2007; Stroud et al. 2015). Much research on online comments has focused on the problems and benefits of allowing anonymous participation (boyd 2012; Elgesem and Nordeide 2016; Santana 2014). Domingo (2011) identifies two main strategies for managing audience participation in online newspapers: The "playground" strategy, in which participation is relegated to a separate space for free experimentation, or the "source" strategy in which participation is much more closely managed and controlled by editorial staff. Reich (2011), discussing online comments in particular, identifies two main strategies for moderation: "An interventionist strategy [that] insists on pre-moderation of every comment" on the one hand, and "a relatively autonomous strategy of post-moderation" on the other. Relevant to the Nordic context of this study, research conducted after the 2011 terror attack and following backlash against online comments in Norway, shows a tendency towards stricter editorial control (Figenschou and Thorbjørnsrud 2016; Ihlebæk, Løvlie, and Mainsah 2013). In (Løvlie, Ihlebæk, and Larsson 2018) this development is discussed in light of a model for understanding strategies for editorial control with online comments set out on a spectrum between "interventionist" and "noninterventionist" strategies, outlined in Figure 1. The lines in the figure describe some of the main design and policy choices that may be used to impose control. A survey among readers who write online comments suggests that many commenters struggle to understand editorial control measures, and that many respondents have a strong experience of antagonism towards moderators and editors. [INSERT FIGURE 1 ABOUT HERE]
3
Design The growing interest in design among media and communications researchers mentioned in the introduction has developed in parallel with a related shift in design research. Here design scholars have presented approaches which go beyond issues of usability and task accomplishment and focused instead on expressive and experiential qualities, sometimes referred to as “third wave HCI” (Bødker 2006; Harrison, Tatar, and Sengers 2007), experience design (McCarthy and Wright 2004) or “humanistic HCI” (Bardzell and Bardzell 2015). However, as pointed out by Löwgren and Reimer (2013b), approaches founded in interaction design and HCI seem to come up short when approaching the highly participatory mass media applications that dominate the contemporary web – such as social media, discussion platforms and systems for online comments. Löwgren and Reimer suggest that the design of such media should be approached from a perspective that combines interaction design with media and communication studies: “media and communication studies need to embrace an interventionist stance in order to produce meaningful and relevant knowledge” (Löwgren and Reimer 2013b, 98–99). Given the strong societal interest in online comments in recent years, one might expect that creating better systems for online comments would be a welcome challenge for media scholars/designers. However, research on online comments for news media has been dominated by theoretical and empirical studies of existing systems and their use. One notable exception is Birchall and S. Coleman’s (2015) use of political theories about democratic deliberation to formulate five "technical considerations" for "deliberative design": • • • • •
Balancing between appealing to the commenters’ passions, and encouraging "some degree of dispassionate rationality" Synchronicity: Using real-time interaction to "replicate the vivacity of face-to-face interactions", or asynchronous deliberation to promote more reflective debate? Visualizing arguments, in order to "level the point of entry to deliberation" Moderation practices: Filtering content or facilitating and directing the conversation? Whether to require user authentication, which may facilitate trust among participants but also sets up a barrier to participation (Birchall and Coleman 2015)
Referring back to the distinction made in the introduction between design as a set of features and as a process, Birchall and Coleman's list is clearly directed at design features. Turning our attention to design processes, one issue merits particular attention: The role of the users. While the development of participatory formats has garnered much attention in the news industry, there is a great difference between the professional norms of journalists and designers regarding user involvement. In spite of the many participatory experiments that have taken place both in the news media and in journalism research and education, audience involvement in mainstream news production remains an exceptional or experimental activity, and does not occupy anything resembling the prominent position that participatory and user-centric approaches hold in the design disciplines. As an example, consider the "epiphanies" of a group of BBC1 executives when they first tried out a user-centered innovation method: "The first [epiphany] was that they did not really understand their customers" (Carlson and Wilmot 2006, 104–5). Hedemann, writing about his efforts in championing the same user-centered method in NRK2, recounts the resistance he often has
1
British Broadcasting Corporation.
2
Norwegian Broadcasting Corporation.
4
encountered from colleagues: "Historically, the task of journalists has not been to understand the audience’s reality. The task has been to make reality understandable for the audience. [...] One of the most common objections I hear is that we shouldn’t leave it to the audience to figure out what stories we should tell" (Hedemann 2010, 34). Singer et al. (2011) have documented that similar attitudes are widespread even in newsrooms experimenting with participatory formats, succinctly summarized by Reich: "As a practical matter, journalists always have preferred their audiences “imagined”" (Reich 2011, 99). On the other hand, user involvement is doxa in the design disciplines, where standard textbooks often recommend extensive user involvement throughout the various stages of the development process, from research through ideation, prototyping and testing (see for instance Hartson and Pyla 2012; Rogers, Sharp, and Preece 2011; Beyer 2010). Compared to these processes, Hedemann’s description of how NRK developed their production of the 2010 Eurovision finale – aiming for an audience of 125 million – based on an interview with one single viewer may seem quite haphazard. As discussed elsewhere, the differences in approach between designers and media professionals may be explained by the logic inherent in the media professions, in particular the high tempo of production in news journalism, as well as the more linear processes of media production when compared to design (Løvlie 2016; Karlsen and Løvlie 2017). However, considering the development of systems for online comments, another important difference between the design fields and the news media may lie in the scope of the user community. Whereas participatory design approaches may focus on communities of interest and the formation of small-scale "publics" around particular issues (Dantec and DiSalvo 2013), when a public broadcaster tries to facilitate political debate "the public" is usually not seen as a niche community but rather the entire citizenry, either on a local or a national scale. In fact, the ability to provide a shared platform which is viewed/used by a majority of the citizens is often cited as one of the chief values of public broadcasting. In design, it is a commonsensical observation that it is much more complicated to design a system aimed at "everyone and no one in particular" rather than one that is tailored to a specific community and use. As such, one may need to consider whether there are complications inherent in this challenge that are not fully taken into account by common approaches to the design of platforms for smaller communities. However, the important task often referred to as "infrastructuring" (Karasti and Syrjänen 2004) is likely to be an important challenge also when designing an online comments system for a public broadcaster. Löwgren and Reimer, summing up recent trends in interaction design on this issue conclude that infrastructure design should not be viewed as a linear and clearly delimited process, but rather as "an ongoing process, interwoven with use... that design not only precedes production and consumption, but rather that infrastructures are constantly evolving" (Löwgren and Reimer 2013b, 146–47). Interestingly, Birchall and Coleman note the focus on small-scale communities as a limitation with constructive research in online deliberation, which has tended to take the form of experiments with "niche products, operating within realms of specific consultative environments, rather than reaching out to the general public" (Birchall and Coleman 2015, 277; see also Davies and Gangadharan 2009). They suggest future research "might involve the development of hybrid spaces of deliberation, in which mass-media audiences are encouraged to go online and participate in debates triggered by television stories and images" (Birchall and Coleman 2015, 277). The case studied in this article fits the description above almost perfectly, and may hopefully contribute to understanding the challenges inherent in facilitating online debate for a broad audience. Method This article reports on a production study based on interviews and observations with employees of the Danish Broadcasting Corporation. The data collection spans a period of three years, offering a
5
longitudinal perspective. The observations and interviews were for the main part conducted by me, while the interviews were transcribed by a research assistant. In line with ethnographic research practice, I attempt to make my own position clear in relation to the object of study, and refer to myself in the first person. I first came in contact with the team developing the new DR Debat website in December 2014 when the team leader invited me to share insights from academic research with the rest of the team. After this meeting, we agreed that I could follow the project doing observations and interviews. Data collection took place primarily at four points in time: First, an interview with two core team members in June 2015, prior to the launch of the website. Second, I observed the first debate organized by the team on 3 September 2015, after which I did a follow-up interview. Third, I returned nearly half a year later in February 2016 to do a new interview and observe two new debates, on 11 and 18 February. Finally, I returned once again in November 2017 in order to gather information about a new development in the project, conducting two new interviews. My analysis is also based on data from an internal evaluation report from DR’s research department, as well as other data collected from the website such as screenshots and html files. During observation, I was mostly a passive observer taking notes, sitting on a desk next to the team and occasionally asking clarifying questions. Audio was recorded during the interviews and two of the three observation sessions, and transcribed for analysis. The first observation session was not recorded because it took place in an open office landscape where it was impossible to establish informed consent from everyone present, so analysis of this session is based on notes and saved materials from the debate website (screenshots and html files). As will be discussed further on, the project experienced a number of changes in personnel and leadership through the period of data collection. I will refer to the informants using pseudonyms because many of them are no longer involved in the project and do not wish to be seen as speaking on behalf of the team. All the informants have signed informed consent forms and have approved the quotes used in the article. I have also hidden the real names of ordinary commenters where they appear in figures. Some professionals appearing in the figures (e.g. the TV show host) have not been anonymized because they are professionals who are intentionally making public statements. List of informants, with aliases and professional role in the project: • • • • • • •
Alex, editor/concept developer (team leader from mid-2015) Betty, team leader (until spring 2015) Cindy, commissioning editor Dennis, journalist/digital editor Elisabeth, journalist/digital editor Frank, journalist/digital editor George, web developer
Case: dr.dk/debat The main purpose of the dr.dk/debat project, according to my respondents, was to improve the quality of debate at the DR website and help fulfil DR’s public service commitment to facilitate public debate. In my interviews with the team members, they often struggled to explain in any further detail the ambitions for the project. Alex repeatedly insisted that their goal was only to improve the quality of debate, not to generate large volumes of traffic, brand loyalty or other forms of commercial goals. He explained the idea by invoking a negative comparison with other websites, in particular in the tabloid press:
6
On many of the debate sites you find a very hard tone, many ad hominem attacks and hard language. Our idea was that we would like to raise the bar, so first of all the site should be a pleasant place to be, you shouldn't feel that you would all the time be personally attacked, but also that you would learn something. 3
However, Cindy – the commissioning editor – considered that one of the greatest challenges for the project was in fact to reach a broad audience: "If we do all those classic, serious political debates, then obviously we are addressing a particular audience. [...] I think we might open up for some other audiences who might find their way to us and discover that this is a pleasant place to learn something." In the following, I will present my findings relating to the two perspectives on design I have outlined earlier: design as a set of system features, and design as a process. Design features From the beginning, the dr.dk/debat website was designed to emphasize stronger editorial control than previous commenting systems at DR. The site would be open for commenting only at particular times, usually in connection with a debate show on TV. Debates would be moderated by a journalist acting as “host”, and would also include a panel of invited experts or prominent figures, who would be visibly present on the website (see Figure 2). In order to participate in the debate, users would have to register and agree to have their real name posted along with their comments. At the start of the debate the website would contain a series of articles giving background information, posted as a "snippet": an item with a headline, an image, a lead text and a link to the main article. User comments would be posted in threads forming under each snippet, in a format similar to the common convention on social media like Facebook. Users could respond to comments in sub-threads, and they could also vote on comments by clicking on links saying "agree" or "don't agree" (Figure 3). Particularly interesting comments from users could also be extracted from these threads by the debate host and turned into new snippets that would move to the top of the page and allow new comment threads to grow under them – thus setting up a mechanism for promoting contributions that the editorial team found particularly noteworthy (Figure 4). [INSERT FIGURE 2 ABOUT HERE] [INSERT FIGURE 3 ABOUT HERE] [INSERT FIGURE 4 ABOUT HERE] My observations of the team’s work cover the period from the site’s launch in September 2015 until February 2016. In this period, the website developed through several changes. After some experiments, the team stopped running debates at other times than directly during and after an accompanying TV broadcast, because they would not get enough activity at other times. They also stopped writing background articles to serve as starting points for debate, and highlighting selected user comments. This latter change was done to simplify the system both for the users and the hosts, because usability issues made it difficult to keep track of the debate. Furthermore, team members speculated that a feature lacking in the user interface had a limiting effect on the debate: there was no function for notifying participants if someone responded to their comments or mentioned them in another thread. Therefore, participants would only be able to engage in a dialogue with each other if
3
All quotes from interviews and observations are my translations from Danish.
7
they stayed on the site, monitoring the threads they commented. Perhaps unsurprisingly, this appeared to happen very rarely. In interviews, team members described the development of dr.dk/debat as a qualified success, in particular regarding the low number of comments they needed to remove. However, an internal evaluation report written in October 2015 shows that the website struggled to gather traffic, with daily comments ranging from 28 to 254, and daily users ranging from 600 to 6200 in the first two months. In the fall of 2017, the website was closed. According to Cindy, the decision to close the website was taken due to persistent problems regarding the usability of the site, along with technical challenges regarding both the login system and the ability to handle large numbers of users during live debates. The broadcaster had now decided to commission a new debate module based on a chat format called Tæt På ("Up Close", my translation), which had recently been developed to facilitate Q&A-sessions with online audiences (Figure 5). Cindy's explanation made it clear that both technical considerations and appeal to users played a role: Tæt På works really well, rarely crashes, is apparently quite easy for the users and is very, very popular. After a referral from TV to a Tæt På session we can easily get anywhere between 600 and 1600 questions in 45 minutes, while in the same time at dr.dk/debat we could struggle to get 10–20 people to participate in a debate.
As the broadcaster was planning some adjustments to the Tæt På system, they realized that it could be expanded to also facilitate a debate format. At the time of writing this article (December 2017), the expanded Tæt På system is still in development with launch planned in the spring of 2018. According to Cindy, the new system will have some notable differences compared to dr.dk/debat: There will be a smaller panel of 1–2 professional participants, with an invisible moderator who premoderates questions and comments from online users before passing them to the panel participants. The system also includes a video module, so that the invited participants could be debating live in front of a camera while reacting to questions coming in from viewers. But in the new system the interaction with users will be constrained to the question and answer format: there is no functionality for users to respond to comments from other users. [INSERT Figure 5 ABOUT HERE]
Design process The dr.dk/debat project started in 2013 when the external design bureau Sorthvid was commissioned to develop a new concept for online debate. They delivered a concept inspired by the popular discussion site reddit.com. The project was then taken over by developers in DR’s technical web development department, working alongside an editorial team. The technical team consisted of 10 people including developers, one designer and a project leader. They worked on the project up to the launch of the website, after which the editorial team took over the main responsibility for the site. After the closure of the website, the editorial team has been disbanded and the development of the new Tæt På system has been a small-scale collaboration between Cindy, George and two of George’s colleagues in the technical department. As my data collection started quite shortly before the public launch of the dr.dk/debat website, my information about the development process is mainly based on interviews in which the participants looked back at the process. Their descriptions reveal that the development process was
8
seen as difficult by many of the stakeholders. Two aspects of this process are particularly interesting here: the process model and the absence of any testing or involvement of external users. Regarding the process model, the technical department preferred to work according to the widespread agile software development methodology SCRUM (Schwaber and Beedle 2002), which predicates an iterative, non-linear approach. However, Cindy revealed that the overall process had in fact been quite linear: "The debate site was to a large degree built from an early vision where someone had drawn up what it should look like, and then that was what we built." Among proponents of agile software development, some of the main criticisms of linear approaches is that they risk resulting in products that lack robustness and are poorly aligned with user needs, and that the cost of changes to the design increases rapidly throughout the project. All three of these problems appear to have occurred in the development of dr.dk/debat, and when the problems became clear after the launch of the site, there were no resources left to deal with them: That whole agile process, making a plan that says we should save some resources to adjust later, that wasn’t done with the old debate site. When the site launched and after one month we could see that this doesn’t work the way we had hoped, it was difficult to prioritize resources to further develop the system. (Cindy)
Cindy describes the development of Tæt På as a paradigm shift, embracing the agile approach: "By and by, we have realized that the smart thing is to build it bit by bit along the way, and then test it, see if it works and then build further from that." However, in spite of the reference to testing, this shift does not seem to entail a design process that involves external users. Both Cindy and George confirmed that there was no user involvement of any kind in the development of the dr.dk/debat website – not even in the form of user testing. When I asked George about how the systems were tested, he explained in some detail about automated technical testing of components in the system, but no testing was conducted on human users outside the technical team. Similarly, when I asked Cindy about user testing she referred to a usability evaluation of the dr.dk/debat website conducted in October 2015 – more than a month after the launch of the website. Cindy also revealed that no user testing was planned for the new system either: We could certainly test on some users before we launch. But whether we are doing that in this case, I don't think so. I think we feel fairly confident in just getting it out and start using it, live, and then adjusting and learning from that. Because we know that the technical bit works, and it’s close to what we already know from the chat module with just a few adjustments. (...) We will discover quickly how it works, and if we can’t control it we will need to quickly figure out what to do.
According to George, this lack of user testing is in fact not unusual for DR; testing on external users is only routinely done in larger projects, such as the development of the broadcaster's video streaming application. Discussion and conclusion While the DR team put much emphasis on the design features of their debate system, they seem to have paid comparatively less attention to the design process – and in particular to user involvement in the process. This lack of attention to the users seems to have contributed to the project’s problems. If we consider the dr.dk/debat project in light of the model for editorial control presented in Figure 1, it is clear that the team had opted for relatively strong control on most dimensions. The website required users to identify with their real names, the moderators were active and clearly visible on the site, the topic for the debate was defined by the hosts, and the debate was open only for 9
a limited amount of time. Overall, the DR Debat team had chosen an unusually interventionist strategy for editorial control. Increasing editorial control runs the risk of setting up barriers to participation (see Birchall and Coleman 2015; Elgesem and Nordeide 2016; Reich 2011), and the website seems to have suffered from this. However, the Tæt På format that has now been chosen to replace the older system represents a further tightening of control on many dimensions: While the demand for commenters to authenticate is being dropped, the broadcaster instead has chosen to introduce pre-moderation, effectively removing the possibility that users may post anything deemed unacceptable by the moderators. The time restrictions will also be further tightened – "one hour maximum", according to Cindy. Furthermore, the question/answer format is likely to ensure that the moderators will shape the conversation even more actively than before, and the restrictions regarding topic will be determined by the moderators and the invited guests. The Tæt På format represents a highly constrained form of participation, indeed not so far from the letters to the editor or call-in radio shows known from pre-digital times. In light of the strategic models for managing audience participation described by Domingo (2011), the Tæt På system seems to represent a return to the "Sources" model, where editorial control is prioritized in the hope that audience contributions may yield material for new journalistic stories. According to Cindy, the "chats" that are currently run using the Tæt På system are reliable sources of journalistic material "in nine out of ten cases", and for her this is a success criterion for the debate module as well. It is illustrative to regard the design features of dr.dk/debat in light of the five considerations outlined by Birchall and Coleman (2015). The largest challenge seems to have been caused by the last of these considerations: user authentication. For the users, the authentication system appears to have set up a barrier which has contributed to low levels of participation on the site. On the technical side, this system seems to also have had an effect regarding the system’s ability to handle synchronous (live) debates, because challenges associated with the authentication system have contributed to instability and system crashes. Furthermore, regarding the balance between passion and rationality, it would seem that the dr.dk/debat system placed great emphasis on facilitating rational debate, but failed to sufficiently engage the audiences. Cindy speculates that the inclusion of both invited panelists and a debate host may have turned attention away from the ordinary users: There were quite a lot of people who all should say their bit sometimes, a host who should welcome everyone and then 2–3 or even 4 professional debaters who each should get their message out and also comment on each other – and then maybe there wasn't really time left for the users.
In the views of both Cindy and George, the Tæt På system will amend several of these issues: Removing the need for authentication will lower the barrier to participation and help simplify the technical architecture of the system, while the question/answer format will greatly simplify interactions and help users navigate on the site. However, while the quote above seems to indicate a need to give more room (or time) to the users, it seems likely that the Tæt På format will do the opposite. As we have seen above, the dr.dk/debat system made it difficult for commenters to engage in dialogue with other users; and in the new Tæt På system this possibility is effectively removed. These observations point to a need for a sixth dimension in the model in Figure 1: addressing the degree to which commenters are able to engage in dialogue not just with moderators and other "backenders" (cf. Toepfl and Litvinenko 2017), but also with each other. It is central to optimistic views about the democratic potential of online comments that they allow ordinary people not just to voice their opinion, but also to engage in debate and deliberation. In the Coral project’s public "list of aggregate user needs", the first two items on the list read: "Be part of a community" and "A way to talk to each other" (The Coral Project n.d.).
10
If one accepts the argument put forward in the introduction to this article, that constructive journalism should pay attention to audience engagement also in the form of commenting and debate, the case discussed here may carry some important lessons. Given the resistance to user-centric approaches recounted elsewhere by sources in BBC and NRK (Carlson and Wilmot 2006; Hedemann 2010), it seems likely that DR is not alone in resisting user involvement. Taking into account the extensive problems that the DR Debat project has encountered, media professionals would do well to draw lessons from the strong emphasis on process and user involvement in design research. In particular when aiming to develop formats for online comments and debate, it seems important to note that "infrastructuring" should be seen as an ongoing, crucial process that require sustained effort and a strong involvement with the user community. After all, what good would constructive journalism do for society if the constructive angles and solutions it presents do not lead to further engagement among the public, inspiring debate directed at finding solutions to important societal challenges? The increasing concerns about the role of Facebook and other social media platforms in democratic processes and abuse of personal data makes it problematic to suggest that news media should give up on online comments and leave commenting and debate to the large social media systems. Rather, a constructive journalism approach seems to call for continued attention to involving the audience – also in the search for solutions for the problems associated with online comments. References Ahva, Laura, and Mikko Hautakangas. 2017. “Understanding ‘Constructive’ through Action Research.” Conference presentation presented at the Nordmedia 2017: Mediated Realities Global Challenges, Tampere, Finland, August 17. Bardzell, Jeffrey, and Shaowen Bardzell. 2015. “Humanistic HCI.” Synthesis Lectures on HumanCentered Informatics 8 (4): 1–185. https://doi.org/10.2200/S00664ED1V01Y201508HCI031. Beyer, Hugh. 2010. User-Centered Agile Methods. San Rafael, CA: Morgan and Claypool Publishers. Birchall, Christopher, and Stephen Coleman. 2015. “Creating Spaces for Online Deliberation.” In Handbook of Digital Politics, edited by Stephen Coleman and Deen Freelon, 264–80. Northampton, MA: Edward Elgar Publishing. http://dx.doi.org/10.4337/9781782548768.00025. Bødker, Susanne. 2006. “When Second Wave HCI Meets Third Wave Challenges.” In Proceedings of the 4th Nordic Conference on Human-Computer Interaction: Changing Roles, 1–8. NordiCHI ’06. New York, NY, USA: ACM. https://doi.org/10.1145/1182475.1182476. Bolter, Jay David. 2003. “Critical Theory and the Challenge of New Media.” In Eloquent Images: Word and Image in the Age of New Media, edited by Mary E. Hocks and Michelle R. Kendrick, 19–36. Cambridge, MA: The MIT Press. Bolter, Jay David, Maria Engberg, and Blair MacIntyre. 2013. “Media Studies, Mobile Augmented Reality, and Interaction Design.” Interactions XX (1): 36–45. boyd, danah. 2012. “The Politics of ‘Real Names.’” Communications of the ACM 55 (8): 29. https://doi.org/10.1145/2240236.2240247. Canter, Lily. 2013. “The Misconception of Online Comment Threads.” Journalism Practice 7 (5): 604–19. https://doi.org/10.1080/17512786.2012.740172. Carlson, Curtis R., and William W. Wilmot. 2006. Innovation: The Five Disciplines for Creating What Customers Want. New York: Crown Business. Chung, Deborah S. 2008. “Interactive Features of Online Newspapers: Identifying Patterns and Predicting Use of Engaged Readers.” Journal of Computer-Mediated Communication 13 (3): 658–79. https://doi.org/10.1111/j.1083-6101.2008.00414.x.
11
Coe, Kevin, Kate Kenski, and Stephen A. Rains. 2014. “Online and Uncivil? Patterns and Determinants of Incivility in Newspaper Website Comments.” Journal of Communication 64 (4): 658–79. https://doi.org/10.1111/jcom.12104. Coleman, Renita. 2007. “Picturing Civic Journalism: How Photographers and Graphic Designers Visually Communicate The Principles of Civic Journalism.” Journalism 8 (1): 25– 43. https://doi.org/10.1177/1464884907072418. Dahlgren, Peter. 2005. “The Internet, Public Spheres, and Political Communication: Dispersion and Deliberation.” Political Communication 22 (2): 147–62. https://doi.org/10.1080/10584600590933160. Dantec, Christopher A. Le, and Carl DiSalvo. 2013. “Infrastructuring and the Formation of Publics in Participatory Design.” Social Studies of Science 43 (2): 241–64. https://doi.org/10.1177/0306312712471581. Davies, Todd, and Seeta Peña Gangadharan, eds. 2009. Online Deliberation: Design, Research and Practice. Stanford, CA: CSLI Publications. Domingo, David. 2011. “Managing Audience Participation: Practices, Workflows and Strategies.” In Participatory Journalism: Guarding Open Gates at Online Newspapers, edited by Jane B. Singer, Alfred Hermida, David Domingo, Ari Heinonen, Steve Paulussen, Thorsten Quandt, Zvi Reich, and Marina Vujnovic, 76–95. Chichester, UK: Wiley-Blackwell. Elgesem, Dag, and Thomas Vie Nordeide. 2016. “Anonymity and Tendentiousness in Online Newspaper Debates.” In Journalism Re-Examined: Digital Challenges and Professional Reorientations. Lessons from Northern Europe, edited by Martin Eide, Helle Sjøvaag, and Leif Ove Larsen. Bristol: Intellect Books. Fagerjord, Anders. 2015. “Humanist Evaluation Methods in Locative Media Design.” The Journal of Media Innovations 2 (1): 107–22. Figenschou, Tine Ustad, and Kjersti Thorbjørnsrud. 2016. “Disruptive Media Events.” Journalism Practice 0 (0): 1–18. https://doi.org/10.1080/17512786.2016.1220258. Haagerup, Ulrik. 2014. Constructive News. Zürich: Innovatio. http://www.innovatio.de/. Harrison, Steve, Deborah Tatar, and Phoebe Sengers. 2007. “The Three Paradigms of HCI.” In Proceedings of CHI 2007, 1–18. New York: ACM. http://people.cs.vt.edu/~srh/Downloads/HCIJournalTheThreeParadigmsofHCI.pdf. Hartson, Rex, and Pardha Pyla. 2012. The UX Book: Process and Guidelines for Ensuring a Quality User Experience. Waltham: Morgan Kaufmann. Hedemann, Ole. 2010. Idéutvikling i Mediehuset. Fredrikstad: IJ-forlaget. Ihlebæk, Karoline Andrea, Anders Sundnes Løvlie, and Henry Mainsah. 2013. “Mer Åpenhet, Mer Kontroll? Håndteringen Av Nettdebatten Etter 22. Juli.” Norsk Medietidsskrift 20 (3). Karasti, Helena, and Anna-Liisa Syrjänen. 2004. “Artful Infrastructuring in Two Cases of Community PD.” In Proceedings of the Eighth Conference on Participatory Design: Artful Integration: Interweaving Media, Materials and Practices - Volume 1, 20–30. PDC 04. New York, NY, USA: ACM. https://doi.org/10.1145/1011870.1011874. Karlsen, Joakim, and Anders Sundnes Løvlie. 2017. “‘You Can Dance Your Prototype If You like’: Independent Filmmakers Adapting the Hackathon.” Digital Creativity 28 (3): 224–39. https://doi.org/10.1080/14626268.2017.1351992. Kies, R. 2010. Promises and Limits of Web-Deliberation. New York: Palgrave. Koskinen, Ilpo. 2006. “Two Solitudes: Design as an Approach to Media Research.” Nordicom Review 27 (2): 35–47. Krippendorff, Klaus. 2006. The Semantic Turn: A New Foundation For Design. Boca Raton: CRC Press. ———. 2016. “Design.” In The International Encyclopedia of Communication Theory and Philosophy, edited by Klaus Bruhn Jensen, Eric W. Rothenbuhler, Jefferson D. Pooley, and
12
Robert T. Craig, 1–12. Hoboken, NJ: Wiley. https://doi.org/10.1002/9781118766804.wbiect264. Larsson, Anders Olof. 2011. “Interactive to Me – Interactive to You? A Study of Use and Appreciation of Interactivity on Swedish Newspaper Websites.” New Media & Society 13 (7): 1180–97. Lee, Eun-Ju, and Edson C. Tandoc. 2017. “When News Meets the Audience: How Audience Feedback Online Affects News Production and Consumption.” Human Communication Research 43 (4): 436–49. https://doi.org/10.1111/hcre.12123. Løvlie, Anders Sundnes. 2011a. “Locative Literature: Experiences with the Textopia System.” International Journal of Arts and Technology 4 (3): 234–48. https://doi.org/10.1504/IJART.2011.041479. ———. 2011b. “Textopia: Experiments with Locative Literature.” Ph.D. dissertation, Oslo: University of Oslo. ———. 2016. “Designing Communication Design.” Journal of Media Innovations 3 (2). http://dx.doi.org/10.5617/jmi.v3i2.2486. Løvlie, Anders Sundnes, Karoline Andrea Ihlebæk, and Anders Olof Larsson. 2018. “User Experiences with Editorial Control in Online Newspaper Comment Fields.” Journalism Practice 12 (3). https://doi.org/10.1080/17512786.2017.1293490. Löwgren, Jonas, and Bo Reimer. 2013a. Collaborative Media: Production, Consumption and Design Interventions. Cambridge, MA: The MIT Press. ———. 2013b. “The Computer Is a Medium, Not a Tool: Collaborative Media Challenging Interaction Design.” Challenges 4 (1): 86–102. https://doi.org/10.3390/challe4010086. Lunenfeld, Peter. 2004. “Media Design: New and Improved without the New.” New Media & Society 6 (1): 65–70. McCarthy, John, and Peter Wright. 2004. Technology as Experience. Cambridge, MA: The MIT Press. https://www.researchgate.net/publication/224927635_Technology_as_Experience. McIntyre, Karen Elizabeth. 2015. “Constructive Journalism: The Effects of
positive Emotions and Solution Information in News Stories.” Ph.D. Dissertation, Chapel Hill, NC: University of North Carolina at Chapel Hill. Morrison, Andrew. 2011. Inside Multimodal Composition. New York: Hampton Press. Muddiman, Ashley, and Natalie Jomini Stroud. 2017. “News Values, Cognitive Biases, and Partisan Incivility in Comment Sections.” Journal of Communication 67 (4): 586–609. https://doi.org/10.1111/jcom.12312. Nyre, Lars, Solveig Bjørnestad, Bjørnar Tessem, and Kjetil Vaage Øie. 2012. “Locative Journalism: Designing a Location-Dependent News Medium for Smartphones.” Convergence: The International Journal of Research into New Media Technologies 18 (3): 297–314. https://doi.org/10.1177/1354856512441151. Øie, Kjetil Vaage. 2015. “News Narratives in Locative Journalism – Rethinking News for the Mobile Phone.” Journal of Media Practice 16 (3): 245–62. https://doi.org/10.1080/14682753.2015.1116763. Reich, Zvi. 2011. “User Comments: The Transformation of Participatory Space.” In Participatory Journalism: Guarding Open Gates at Online Newspapers, edited by Jane B. Singer, Alfred Hermida, David Domingo, Ari Heinonen, Steve Paulussen, Thorsten Quandt, Zvi Reich, and Marina Vujnovic, Kindle e-book, loc. 2424-3008. Malden, Massachusetts: Wiley-Blackwell. Rogers, Yvonne, Helen Sharp, and Jenny Preece. 2011. Interaction Design: Beyond Human Computer Interaction. Chichester, UK: John Wiley & Sons. Rowe, Ian. 2015. “Deliberation 2.0: Comparing the Deliberative Quality of Online News User Comments Across Platforms.” Journal of Broadcasting & Electronic Media 59 (4): 539–55. https://doi.org/10.1080/08838151.2015.1093482.
13
Ruiz, C., D. Domingo, J. L. Mico, J. Diaz-Noci, K. Meso, and P. Masip. 2011. “Public Sphere 2.0? The Democratic Qualities of Citizen Debates in Online Newspapers.” International Journal of Press-Politics 16 (4): 463–87. https://doi.org/10.1177/1940161211415849. Santana, Arthur D. 2014. “Virtuous or Vitriolic.” Journalism Practice 8 (1): 18–33. https://doi.org/10.1080/17512786.2013.813194. Schwaber, Ken, and Mike Beedle. 2002. Agile Software Development with Scrum. Upper Saddle River, NJ: Prentice Hall. Simon, Herbert. 1969. The Sciences of the Artificial. Cambridge, MA: The MIT Press. Singer, Jane B., Alfred Hermida, David Domingo, Ari Heinonen, Steve Paulussen, Thorsten Quandt, Zvi Reich, and Marina Vujnovic, eds. 2011. Participatory Journalism: Guarding Open Gates at Online Newspapers. Malden, Massachusetts: Wiley-Blackwell. Skjulstad, Synne. 2008. “Mediational Sites: A Communication Design Perspective on Websites.” Ph.D. dissertation, Oslo: University of Oslo. Sobieraj, Sarah, and Jeffrey M. Berry. 2011. “From Incivility to Outrage: Political Discourse in Blogs, Talk Radio, and Cable News.” Political Communication 28 (1): 19–41. https://doi.org/10.1080/10584609.2010.542360. Springer, Nina, Ines Engelmann, and Christian Pfaffinger. 2015. “User Comments: Motives and Inhibitors to Write and Read.” Information, Communication & Society 18 (7): 798–815. https://doi.org/10.1080/1369118X.2014.997268. Stroud, Natalie Jomini, Joshua M. Scacco, Ashley Muddiman, and Alexander L. Curry. 2015. “Changing Deliberative Norms on News Organizations’ Facebook Sites.” Journal of Computer-Mediated Communication 20 (2): 188–203. https://doi.org/10.1111/jcc4.12104. Sunstein, Cass R. 2017. #Republic: Divided Democracy in the Age of Social Media. Princeton and Oxford: Princeton University Press. http://press.princeton.edu/titles/10935.html. The Coral Project. n.d. “Research.” The Coral Project. n.d. https://coralproject.net/research.html. Toepfl, Florian, and Anna Litvinenko. 2017. “Transferring Control from the Backend to the Frontend: A Comparison of the Discourse Architectures of Comment Sections on News Websites across the Post-Soviet World.” New Media & Society, September, 1461444817733710. https://doi.org/10.1177/1461444817733710. Wright, Scott, and John Street. 2007. “Democracy, Deliberation and Design: The Case of Online Discussion Forums.” New Media & Society 9 (5): 849–69. Ziegele, Marc, Mathias Weber, Oliver Quiring, and Timo Breiner. 2017. “The Dynamics of Online News Discussions: Effects of News Articles and Reader Comments on Users’ Involvement, Willingness to Participate, and the Civility of Their Contributions.” Information, Communication & Society 0 (0): 1–17. https://doi.org/10.1080/1369118X.2017.1324505.
14
Figure 1: Strategies for editorial control with online comments. Figure 2: The dr.dk/debat website, with names and photos of the debate host and three debate panelists (3 Sept 2015). (The names and photos of informants have been obscured in order to preserve their anonymity.) Figure 3: Snippet with comment field and one comment. The commenter's name has been blacked out. The labels "enig" (agree) and "uenig" (don't agree) are links that can be clicked to signal agreement with the comment. (Screenshot 3 Sept 2015.) Figure 4: A comment has been extracted and turned into a new "snippet". In this case the original comment was posted by one of the panel members. The names of the other commenters have been blacked out. (Screenshot from 3 Sept 2015.) Figure 5: A question-and-answer session in the Tæt På format from dr.dk 13 October 2017.
15
